Expressive Time Series Querying with Hand-Drawn Scale-Free Sketches

@article{Mannino2018ExpressiveTS,
  title={Expressive Time Series Querying with Hand-Drawn Scale-Free Sketches},
  author={Miro Mannino and Azza Abouzeid},
  journal={Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems},
  year={2018}
}
  • Miro Mannino, A. Abouzeid
  • Published 21 April 2018
  • Computer Science
  • Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems
We present Qetch, a tool where users freely sketch patterns on a scale-less canvas to query time series data without specifying query length or amplitude. We study how humans sketch time series patterns --- humans preserve visually salient perceptual features but often non-uniformly scale and locally distort a pattern --- and we develop a novel matching algorithm that accounts for human sketching errors. Qetch enables the easy construction of complex and expressive queries with two key features… 

Figures and Tables from this paper

Qetch: Time Series Querying with Expressive Sketches
TLDR
Qetch is designed, a query-by-sketch tool for time series data, top-down that develops a matching algorithm that accounts for human sketching errors and enables the easy construction of expressive queries that include regular expressions over sketches and queries over multiple time series.
A Demonstration of Scale-Free Query by Sketching with Qetch
TLDR
Qetch, a time series querying tool, where users freely sketch patterns of interest, without specifying query length or amplitude, on a scale-free canvas is demonstrated.
Sketch-Based Fast and Accurate Querying of Time Series Using Parameter-Sharing LSTM Networks
TLDR
This article introduces a machine learning based solution for fast and accurate querying of time series data based on a swift sketching interaction that builds on existing LSTM technology (long short-term memory) to encode both the sketch and the time seriesData in a network with shared parameters.
From Sketching to Natural Language
TLDR
Motivated from the findings from Zenvisage, ShapeSearch is developed, an efficient and flexible pattern-searching tool that enables the search for desired patterns via multiple mechanisms: sketch, natural-language, and visual regular expressions.
From Sketching to Natural Language: Expressive Visual Querying for Accelerating Insight
TLDR
This work develops an efficient and flexible pattern-searching tool that enables the search for desired patterns via multiple mechanisms: sketch, natural-language, and visual regular expressions, and supports query-aware and perceptually-aware optimizations to execute shape queries within interactive response times.
ShapeSearch: A Flexible and Efficient System for Shape-based Exploration of Trendlines
TLDR
This work proposes ShapeSearch, an efficient and flexible pattern-searching tool that enables the search for desired patterns via multiple mechanisms: sketch, natural-language, and visual regular expressions, and develops a novel shape querying algebra, with a minimal set of primitives and operators.
You can't always sketch what you want: Understanding Sensemaking in Visual Query Systems
TLDR
The study results reveal that ad-hoc sketch-only querying is not as commonly used as prior work suggests, and suggest that all three sensemaking processes must be integrated in order to make future VQSs useful for a wide range of analytical inquiries.
Qute: Query by Text Search for Time Series Data
TLDR
Qute (Query by Text) is a text-based search which leverages information retrieval features such as relevance feedback and subsumes motif and discord/anomaly discovery.
PSEUDo: Interactive Pattern Search in Multivariate Time Series with Locality-Sensitive Hashing and Relevance Feedback
We present PSEUDo, an adaptive feature learning technique for exploring visual patterns in multi-track sequential data. Our approach is designed with the primary focus to overcome the uneconomic
Progressive Similarity Search on Time Series Data
TLDR
This paper presents preliminary ideas on computing probabilistic estimates of the final results that could help users decide when to stop the search process, i.e., deciding when improvement in the final answer is unlikely, thus eliminating waiting time.
...
1
2
3
4
...

References

SHOWING 1-10 OF 47 REFERENCES
How do humans sketch objects?
TLDR
This paper is the first large scale exploration of human sketches, developing a bag-of-features sketch representation and using multi-class support vector machines, trained on the sketch dataset, to classify sketches.
Relaxed selection techniques for querying time-series graphs
TLDR
Relaxed selection techniques are introduced, in which users implicitly define a level of similarity that can vary across the search pattern, while creating a search query with a single-gesture interaction.
Sketching a graph to query a time-series database
TLDR
QuerySketch, a financial database application in which graphs are used for query input as well as output, allows users to sketch a graph freehand, then view stocks whose price histories match the sketch.
RINSE: Interactive Data Series Exploration with ADS+
TLDR
RINSE is introduced, a system that allows users to experience the benefits of the ADS+ adaptive index through an intuitive web interface and can scale to large data sizes, while drastically reducing the data to query delay.
Evaluating Subjective Accuracy in Time Series Pattern-Matching Using Human-Annotated Rankings
TLDR
This paper describes how a human-annotated ranking based on real-world queries and datasets can be created using simple crowdsourcing tasks and uses this ranking as ground-truth to evaluate the perceived accuracy of existing time series similarity measures.
Dynamic Query Tools for Time Series Data Sets: Timebox Widgets for Interactive Exploration
TLDR
Extended case studies involving the analysis of two different types of data from molecular biology experiments provided valuable feedback and validated the utility of both the timebox model and the TimeSearcher tool.
QueryLines: approximate query for visual browsing
We introduce approximate query techniques for searching and analyzing two-dimensional data sets such as line or scatter plots. Our techniques allow users to explore a dataset by defining QueryLines:
Efficiently Finding Arbitrarily Scaled Patterns in Massive Time Series Databases
TLDR
A novel technique to speed up similarity search under uniform scaling is demonstrated, which can achieve a speedup of 2 to 3 orders of magnitude under realistic settings.
Warping indexes with envelope transforms for query by humming
TLDR
This work treats music as a time series and exploit and improve well-developed techniques from time series databases to index the music for fast similarity queries and improves on existing DTW indexes technique by introducing the concept of envelope transforms, which gives a general guideline for extending existing dimensionality reduction methods toDTW indexes.
PhotoSketch: a sketch based image query and compositing system
TLDR
A system for progressively creating images through a simple sketching and compositing interface that allows generating complex images in an intuitive way is introduced.
...
1
2
3
4
5
...